The rates, per card
Every card in a pack is rolled on its own, so these are per-card chances, not per-pack chances. Each tier then draws a random card from its own slice of the 25-card set.
| Rarity | Chance per card | Cards in the set | Points per pull |
|---|---|---|---|
| Common | 60% | 10 | 10 |
| Uncommon | 25% | 6 | 25 |
| Rare | 10% | 4 | 60 |
| Epic | 4% | 3 | 150 |
| Legendary | 1% | 2 | 400 |
What those rates mean for a pack
A pack is five independent rolls, so the per-pack numbers are not simply five times the per-card numbers. The correct way to work them out is to take the chance of missing on every card and subtract it from one: for a Legendary, that is 1 - 0.99^5 = 4.9%.
| What you want from one pack | Odds | Working |
|---|---|---|
| At least one Rare or better | 55.6% | 1 - 0.85^5 |
| At least one Epic or better | 22.6% | 1 - 0.95^5 |
| At least one Legendary | 4.9% | 1 - 0.99^5 |
| All five Commons | 7.8% | 0.6^5 |
| At least one Legendary in ten packs | 39.5% | 1 - 0.99^50 |
So a pack with nothing above Rare is the less common outcome, not the normal one: more than half of all packs contain at least one Rare or better. On average a pack returns 141.25 points (5 x 28.25, where 28.25 is the expected value of a single card).
The flip side is worth stating plainly. Even at 39.5% across ten packs, roughly three players in five who open ten packs will still not see a Legendary, and a Legendary is worth 400 points -- eight full packs of nothing but Commons.
How hard is one specific card?
Owning all 25 cards is the real completionist goal, and the card you are missing is always the hard one. Each tier picks at random within itself, so the chance of one named card is the tier's rate divided by the number of cards in that tier:
| Card you want | Chance per card | Chance in one pack |
|---|---|---|
| One specific Common | 6% | 26.6% |
| One specific Uncommon | 4.17% | 19.1% |
| One specific Rare | 2.5% | 11.9% |
| One specific Epic | 1.33% | 6.5% |
| One specific Legendary | 0.5% | 2.5% |
A specific Legendary is therefore a 1-in-40 packs proposition, and you have two of them to find. That is the honest cost of finishing the set, and it is why the collection counter shows a fraction rather than a nice round reward.
Why these odds are exact when other simulators' are estimates
The Pokemon Company does not publish official per-rarity pull rates for its English booster packs. Simulators that cover real sets therefore work from opening samples: RipMath says on its own page that its figures are estimated and shows a sample size where it has one, PullSim labels its numbers as sampled rates, and sites such as UnpackCard publish guides on how to read such rates critically. That is a sound approach for real sets, and it has an unavoidable margin of error until the sample gets large.
This site is a different case. The table above is not a sample and not an estimate: those are the exact probabilities the game code applies to every card, and they are the same in every session, on every device, for every player. There is no sample size to check because there is no sampling, and no confidence interval to quote. If the numbers on this page ever stopped matching the game, that would be a bug worth reporting on the contact page.
No pity timer, no streak bonus, no lucky time
There is no pity timer here. No guaranteed Legendary after a dry streak, no accumulating "luck" meter, no bonus for opening ten packs in a row, no lucky hour of the day, and no better odds in a private window. Each card is rolled fresh from the same distribution, so pack six hundred has exactly the same 1% per-card Legendary chance as pack one.
Opening more packs raises your chance of seeing a Legendary at some point. That is arithmetic, not a mechanic: it is the accumulation in the table above, and it never makes the next pack more likely than 1% per card. Any site or video that tells you otherwise is selling a superstition, and any simulator that quietly raises your odds after a bad streak is misreporting its own rates.
How to check this yourself
Open ten packs a few times and read the rarity breakdown that Open 10 packs gives you. Fifty cards should average about 30 Commons, 12 or 13 Uncommons, five Rares, two Epics and half a Legendary. Small samples swing a lot -- that is variance, not a rigged game -- but over a few hundred cards the counts should settle close to the table above.
Last updated: 16 November 2026.